Assessing the Context of Permanency and Reunification in the Foster Care System. 2.3 Evaluation of Programs With Specific Foci


Section 3 of this report, Permanency and Reunification Trends in 25 States, describes programs in 12 locations. Section 2, Permanency: A Balancing Act, puts these programs in four categories:

  • Programs for substance abusing parents. Drug courts are a subtype of programs focused on parental substance abuse. An example of a drug court is the Drug Court Recovery Project in San Diego County, California. Drug treatment is provided by several inpatient and outpatient treatment substance abuse centers. Services other than substance abuse treatment are provided through a county case manager. Parents' failure to comply with treatment may result in the court imposing sanctions such as short stays in jail and fines for contempt of court. Continued violations may result in dismissal from the program and evaluation for termination of parental rights. Phase 1 of the program requires a court appearance once a week for 90 days; phase 2 requires a court appearance once every two weeks for 90 days; and phase 3 requires a court appearance once a month for 90 days. An additional feature of the program is the use of "safe houses," transitional housing units for parents and children while parents are undergoing treatment. This is a large program, about 880 cases were being served in July 2000.

    There are also specialized programs for substance abuse outside the framework of the courts. An example is Mothers Making a Change(MMAC) in Cobb and Douglas Counties in Georgia, a comprehensive program that includes community outreach, early intervention, family preservation and support services, and substance abuse treatment services. The program involves an inpatient phase of 90 days, followed by 9 months of services in the community during which children should be reunified. After that, clients receive 6 months of aftercare services. At the time we talked with this program there were 21 parents in the residential phase and 30 in the nonresidential phases.

  • Programs for children with severe problems. These programs are focused on children with severe mental health problems and are sometimes characterized by "wraparound" services, an effort to bring to bear a comprehensive set of community services to meet child and family needs. Examples are the Santa Clara County, California wraparound services and Boston's Family Reunification Network. Santa Clara County's program accepts children who are in high-end residential care. It is conducted by two private agencies and has the capacity to serve 195 children. Boston's program is conducted by a network of private agencies and tries to reunify children who are in specialized foster care or residential treatment with their parents. The Heller School at Brandeis University conducts research on the program.
  • Community-based programs. Community based approaches to child welfare are being tried in a number of communities. The Family to Family initiative of the Annie E. Casey Foundation is a model that is being used in a number of locations. The Lucas County, Ohio Community Development Department uses the Family to Family model to help facilitate reunification. Six neighborhood community centers provide space for visitation, parenting classes, and family conferences. These multi-service agencies also provide recreation, family counseling, health services, and after-school programs. A Community Advocate program provides intensive in-home services when the child is ready to return home. The county has about 500 children in foster care and 500 in relative care. There are between 80 and 100 families in the Community Advocates program.
  • Family preservation model intensive services. These programs use family preservation like services, often adapting the Homebuilders model. Caseloads are small (2-5 families), services are intensive and short term. An example is the New Jersey Natural Parent Support Program in which families receive services for 6 months. The program is fairly small, each of six counties serve 9 families at a time.

All of the above initiatives deserve evaluation. Of particular urgency is the evaluation of drug courts, both because of the centrality of substance abuse as a problem in child welfare and because drug courts represent an increase in responsibilities and influence of the courts in the child welfare system. An assessment of the effects of strong court sanctions is also needed. We understand that an evaluation of the San Diego program is contemplated and that effort should be closely watched.

Other programs focused on substance abuse should also be evaluated. The Georgia Mothers Making a Change effort could be explored further for possible evaluation. A major issue here is program size. The numbers of cases involved at present may not be sufficient to allow for rigorous evaluation. It is also not known whether there are cases referred that cannot be served because of program capacity limits, thus allowing for random assignment or the construction of a comparison group by other means.

Also much needed are evaluations of community based programs. Some are underway. As mentioned above, Chapin Hall is engaged in an evaluation of the Edna McConnell Clark Foundation's CPPC program. The Casey Foundation's Family to Family programs have research components (for example, the Child Welfare Research Center at the University of California, Berkeley is working with the Los Angeles program). To our knowledge, the Lucas County program is not yet being subjected to rigorous evaluation and could be explored further as an evaluation site.

A major difficulty in evaluating community programs is that they are intended to have community-wide impacts or to involve all cases in a community. (The Lucas County program includes all children in care in the county.) Therefore, forming comparison groups is quite problematic, since one cannot identify contemporaneous cases in the same community that are not affected by the initiative. In this situation, comparisons over time are often suggested (comparing outcomes before the program was implemented) but such comparisons are always subject to alternative explanations of the source of the change (economic conditions are always changing and communities do not otherwise stand still). Comparisons may also be made with other communities in which the initiative is absent but such comparisons are also usually flawed, since one cannot be sure the communities were similar before implementation of the initiative or that their trajectories of change would have been the same without the program (communities are usually chosen for a program because of the likelihood of success). It would be possible to randomly assign communities to interventions but so far foundations do not appear to have taken that option seriously (most community intervention efforts have been supported by foundations).

Programs for children with severe mental health needs and wraparound programs should be evaluated. As we note above, the Boston Family Reunification Network has associated research activities. The possibility of an evaluation of the Santa Clara County program should be explored further. It appears that program may be marginally large enough to support rigorous evaluation, but it is not known whether saturation has been reached.

While it would be desirable to conduct rigorous evaluations of family preservation-like reunification programs, it appears to us that research implementation issues may often be quite daunting. In particular, the programs we are aware of are generally quite small, too small to allow for the accumulation of enough cases within a reasonable period of time.

Given the increased interest in subsidized guardianship as a component of permanency efforts, it is important that evaluative information be developed. Fortunately, a large scale randomized experiment on the Illinois subsidized guardianship program is being conducted by Westat for the State of Illinois. Interim findings from that study showed that the introduction of guardianship as a permanency option increased the number of children placed in permanent arrangements without detracting from the number of children adopted. Questions still to be answered include the continued impact of guardianship on permanency rates and the effects of guardianship on child well being.